Thanks to a generous philanthropic grant (pending final logistics) from Coefficient Giving, 𝘎𝘦𝘰𝘥𝘦𝘴𝘪𝘤 𝘪𝘴 𝘩𝘪𝘳𝘪𝘯𝘨 𝘔𝘦𝘮𝘣𝘦𝘳𝘴 𝘰𝘧 𝘛𝘦𝘤𝘩𝘯𝘪𝘤𝘢𝘭 𝘚𝘵𝘢𝘧𝘧. Come build the base of alignment with us 🤖 Applications now open: airtable.com/appuugUGFPJE...
Kyle O’Brien
@kyletokens.bsky.social
AGI Alignment & Control Pretraining @ Geodesic Research | https://kyobrien.io
Geodesic mission is to develop the science of providing robustly aligned initializations for RL, where alignment priors persist through the remainder of training. Do considering applying if you want to help make sharing the world with superintelligence go well for humanity.
Geodesic Research@geodesicresearch.bsky.social · 3mo agoGeodesic is hiring Members of Technical Staff. We're a Cambridge-based AI safety org. Our seminal work showed you can bake alignment priors into base models. Now, we want to make base models robust to the adversarial effects of long-horizon capabilities RL. EOI ~5 mins: tally.so/r/vG4G6A
Very excited to see pretraining safety efforts! We’re only now beginning to understand how promising pretraining safety and alignment interventions are. Much in the way that curating the base model is important for capabilities like reasoning, so too might it be important for safety.
Katherine Lee@katherinelee.bsky.social · 8mo agoHere's our job posting for more info! jobs.ashbyhq.com/openai/d829b... Please tell me a little about yourself when you email!
I've joined Geodesic Research to build the open-science field of AI safety pretraining research. Our first paper is wild. TL;DR — LLMs pretrained on data about misaligned AIs themselves become less aligned. Luckily, pretraining LLMs with data about good AIs helps them become more aligned.
Alignment Pretraining: AI Discourse Causes Self-Fulfilling (Mis)alignment
LLMs trained on data about misaligned AIs themselves become less aligned. Luckily, pretraining LLMs with synthetic data about good AIs helps them become more aligned. These alignment priors persist th...
alignmentpretraining.ai
The first pretraining results are in, and it looks like models indeed have self-fulfilling misalignment properties. Great work by Tice et al! alignmentpretraining.ai
Alignment Pretraining: AI Discourse Causes Self-Fulfilling (Mis)alignment
LLMs trained on data about misaligned AIs themselves become less aligned. Luckily, pretraining LLMs with synthetic data about good AIs helps them become more aligned. These alignment priors persist…
alignmentpretraining.ai
You know you're AGI-pilled when your Spotify Wrapped looks like this.
Applications to apply for the ERA:AI Fellowship close November 3rd! Participating in this Summer's fellowship was my gateway into pursuing AGI safety research full-time. I will be a research manager for the upcoming Winter fellowships. Feel free to DM me with questions. :) erafellowship.org
ERA Fellowship
ERA is a talent programme supporting early-career researchers and entrepreneurs to understand and mitigate risks from frontier AI, based at Cambridge, UK.
erafellowship.org
📌📌📌 I'm excited to be on the faculty job market this fall. I just updated my website with my CV. stephencasper.com
Stephen Casper
Visit the post for more.
stephencasper.com
Thanks to @stellaathena.bsky.social for chatting with me about Deep Ignorance: the new paper/project from Eleuther AI and the UK AISI. Bottom line: Worried AI could teach people to build bioweapons? Don’t teach it how fortune.com/2025/08/14/w...
AI safety tip: if you don’t want it giving bioweapon instructions, maybe don’t put them in the training data, say researchers
New research shows that scrubbing risky material from AI training data can build safeguards that are harder to bypass — and one author calls out tech giants for keeping such work under wraps.
fortune.com
Big and True :)
Cas (Stephen Casper)@scasper.bsky.social · 12mo ago🧵 New paper from UK AISI x @eleutherai.bsky.social rai.bsky.social that I led with @kyletokens.bsky.social y.social���: Open-weight LLM safety is both important & neglected. But filtering dual-use knowledge from pre-training data improves tamper resistance *>10x* over post-training baselines.
I like that OpenAI published this. They were able to fine-tune away GPT-oss's refusal, decreasing refusal rates to ~0%. These results aren't surprising. Acknowledging that existing safeguards don't generalize to open models is the first step in developing solutions. arxiv.org/abs/2508.031...
Estimating Worst-Case Frontier Risks of Open-Weight LLMs
In this paper, we study the worst-case frontier risks of releasing gpt-oss. We introduce malicious fine-tuning (MFT), where we attempt to elicit maximum capabilities by fine-tuning gpt-oss to be as ca...
arxiv.org
I've learned a lot over the past two years of getting into research, mostly from mistakes. I’ve made many mistakes. Such is science. Good research is often at the adjacent possible. I've written up much of what I've learned now that I'm beginning to mentor others. open.substack.com/pub/kyletoke...
Don’t "Think", Just Think
Lessons From Breaking Into AI Research
open.substack.com
I led an effort at Microsoft last Fall that studied whether SAE steering was an effective way to improve jailbreak robustness. Our paper on SAE steering has been accepted to the ICML Actionable Interpretability Workshop! Venue: actionable-interpretability.github.io Paper: arxiv.org/abs/2411.11296
Steering Language Model Refusal with Sparse Autoencoders
Responsible deployment of language models requires mechanisms for refusing unsafe prompts while preserving model performance. While most approaches modify model weights through additional training, we...
arxiv.org
I'll be in England this summer as an AI Safety Research Fellow with ERA! erafellowship.org/fellowship I will be studying data filtering and tamper-resistant unlearning for open-weight AI safety so that the community can continue to benefit from open models as capabilities improve.
Fellowship — ERA Fellowship
erafellowship.org